J. Agric. Food Chem. 1982, 30, 1133-1135
1133
5-, 6-, and 7-Methoxybenzoxazolinone: Carbon-13 Nuclear Magnetic Resonance Spectra and Biological Activity Gilles F. Nicollier, Paul A. Hedin,* and Frank M. Davis
For elucidation of structureactivity relationships, 5-, 6-, and 7-methoxybenzoxazolinones were synthesized and found to have generally similar activities as larval growth inhibitors of the Southwestern corn borer (Diutraea grandiosellu, Dyar), the fall armyworm (Spodopterafrugiperda, J. E. Smith), and the European corn borer (Ostrinia nubilalis, Hubner). The compounds were also evaluated for inhibition of radish (Raphunussatiuus) and tomato (Lycopersicon esculentum) seedling growth and for antibacterial activity with Pseudomonas maltophilia. Assignments of 13C magnetic resonances (13C NMR) were made by using gated decoupling spectra.
6-Methoxybenzoxazolinone(6-MBOA) has been found in cereal (Virtanen and Hictala, 1955; Gahagan and Mumma, 1967; Venis and Watson, 1978) and in Scopara dulcis (Chen and Chen, 1976). It was reported as a host plant resistance factor to the European corn borer (Ostrinia nubilalis Hubner) (Klun and Robinson, 1969) and also as having hypotensive activity and antidiabetic potential (Venis and Watson, 1978). Synthesis and lH NMR data have been reported for 6-MBOA by Matsumoto and Okazawa (1974), Richey, et al. (1976), and Chen and Chen (1976). The procedure of Matsumoto and Okazawa (1974) was used for our synthesis of 5-, 6-, and 7-MBOA (Figure 1). We discuss the 13C NMR spectra of these three compounds; the attribution of the different shifts to the individual carbons was achieved with gated decoupling used in conjunction with the pulsed F l '(Fourier transformation) technique to retain the increase in signal to noise produced by the nuclear Overhauser effect (NOE) while at the same time giving a completely coupled spectrum. The three isomers (5-, 6-, and 7-MBOA) were tested as growth inhibitors of radish and tomato seedlings, for antibacterial activity with Pseudomonas maltophilia, and for larval growth inhibition of the Southwestem corn borer (SWCB) (Diatraea grandiosella, Dyar), of the fall armyworm (FAW) (Spodopterafrugiperda, J. E. Smith), and of the European corn borer (ECB) (0.nubilalis, Hubner). MATERIALS AND METHODS 'H and 13CN M R were measured at 80 and 20 MHz with tetramethylsilane (Me4&) and 3-(trimethylsily1)-1propanesulfonic acid (DSS) as the internal standards. 'H NMR data are the result of a simple first-order analysis and of comparison with the 6-MBOA of Venis and Watson (1978). Mass spectra were obtained by introduction of the sample into the direct inlet system of the HP 5985 mass spectrometer. Syntheses. Syntheses were performed according to Matsumoto and Okazawa (1974). The general procedures are as follows. Esterification. 3-, 4-, and 5-methoxysalicyclic acids (Aldrich),20 g each, were refluxed in absolute MeOH with Department of Biochemistry, Mississippi State University, Mississippi State, Mississippi 39762 (G.F.N.), Boll Weevil Research Laboratory, US. Department of Agriculture, Agricultural Research Service, Mississippi State, Mississippi 39762 (P.A.H.), and Crop Science and Engineering Research Labroatory, US.Department of Agriculture, Agricultural Research Service, Mississippi State, Mississippi 39762 (F.M.D.). 0021-8561/82/1430-1133$01.25/0
13C NMR Chemical Shift Values (ppm) of 4-, 5-, and 6-MBOA in Me,SO-d, Table I.
7-MBOA
carbons 2 4 5 6 7 8 9 CH,O
Me,SO
CDCI,
5-MBOA 6-MBOA
154.53 102.79 124.35 106.73 143.20 131.76 131.71 56.20
155.77 103.10 124.73 108.12 143.96 132.63 130.96 56.90
155.05 96.46 156.26 106.85 109.79 137.52 131.18 55.74
154.83 109.76 109.12 155.25 97.05 144.14 123.77 55.77
3 g of HC1 (concentrated) for 48 h. The ester was separated from the remaining acid by column chromatography on silica gel (1 = 20 cm, d = 5 cm) with EtOAc/NH,OH (98/2 v/v): yield of esters, 3- = 19.1 g, 4- = 15.7 g, and 5-methyl methoxysalicylate = 17.7 g. Hydrazide. The esters were added (without solvent) to 2.0 mol of hydrazine (Aldrich) at room temperature to form the hydrazides. Heat was evolved and the subsequent reaction mixture crystallized on standing. Excess hydrazine was removed under vacuum with a rotoevaporator: yield of hydrazides, 3- = 14.2 g, 4- = 13.9 g, and 5-substituted hydrazide = 15.0 g. Azide. The hydrazides were dissolved with difficulty by the addition of 20 mL of concentrated HCl, 40 mL of H20,and then an additional 45 mL of HC1 (concentrated). To a chilled solution of the hydrazides (0-5 "C), 14 g of NaN02 in 40 mL of H20, also chilled, was added slowly while maintaining the temperature of 0-5 OC over a period of 15 min. The azide was extracted from the aqueous reaction mixture with CHC1,: yield of azide, 3- = 8.4 g, 4- = 6.9 g, and 5-substituted azide = 10.5 g. MBOA Synthesis. The azides, after evaporation of the CHCl,, were dissolved in dry toluene and refluxed overnight. The reaction product was redissolved in CHC13and chromatographed on a 5 X 20 cm silica gel column, and the MBOAs were eluted with ETOAc/CHCl,, 1/9 (v/v). The final MBOAs were then crystallized and recrystallized from CHC1,. Physical Data. 5-MBOA: yield 8.9 g ; mp 151 "C; MS m / e 165 (M+), 150, 136, 122, 106, 94, 80, 69; 'H NMR (CDC13) 6 3.80 [s, 3 H, OCH&(5)], 6.56-6.67 [d, 1 H, H-C(6), J = 9 Hz], 6.68 [s, 1 H, H-C(4)], 7.04-7.16 (d, 1 H, H-C(7), J = 9 Hz], 9.16 (brs, 1 H, NH); 13C NMR (Me2SO-d6)see Table I. 6-MBOA: yield, 4.9 g ; mp 151 "C; MS m / e 165 (M'), 150, 135,122,107,97, 79,69; lH NMR (CDC13)see Venis and Watson (1978) and Chen and Chen (1976); 13C NMR (Me2SO)see Table I. 7-MBOA: yield, 5.7 g ; mp 144 "C; MS m / e 165 (M+), 150, 122, 107, 0 1982 American Chemical Society
1134
J. Agric.
Food Chem., Vol. 30, No. 6, 1982 R=OCH
in C - 5 5-MEOA in C - 6 6-MBOA in C - 7 7-MBOA
Nicollier, Hedin, and Davis
Table 11. Coupling Constant Values (Hertz) of 5 - , 6 - , and 7-MBOA IJCH
Figure 1. Structure of 5-, 6-, and 7-MBOA. 94, 79, 66; 'H NMR (CDCI,) 6 3.98 [s, 3 H, OCH,-C(7)], 6.68-6.77 [m, 2 H, H-C(6) and H-C(5), JH = 8.2 Hz, JS4 = 1.3 Hz], 7.09-6.99 [d, l H , H-C(4), J = 8 Hz], 9.31 (brs, 1 H, NH); 13C NMR (Me2SO-d6)see Table I. Bioassay Techniques. Seed growth inhibition techniques were used as previously described (Nicollier and Thompson, 1982). The bacterial bioassay with P. maltophilia was performed on a nutrient agar plate. The insect larval growth bioassays were performed by the methods of Davis and Hedin (1982). Newly hatched Southwestern corn borer larvae were placed on a custom Bio-Mix No. 1107 prepared by Bioserv, Inc., Frenchtown, NJ, for 5 days after which they were weighed. The diet was based on wheat germ, casein, sucrose, vitamins, salts, agar, and antimicrobial agents. The candidate compounds (0.002-0.2% of the diet) were either added directly as an aqueous solution or, if required, added in a nonaqueous solvent to the casein and rotoevaporated to remove the solvent. The casein was then mixed with the remainder of the diet. Four repetitions of five insects were employed per data point.
RESULTS AND DISCUSSION The 13C-IH coupling constants, which were obtained from gated decoupling spectra with the aid of chemical shifts, led to the assignment of the resonances of the carbons in 5-, 6-, and 7-MBOA. An improved resolution was obtained upon addition of a small amount of D20, which gave two effects: first, the coupling to the deuteron is about one-sixth of that to the proton; second, deuteron quadrupole relaxation is much faster than proton relaxation. Single resonance spectra may be very valuable in the case of the small-size or even medium-size molecules provided that the spectra are not too complex, but the ambiguity in the assignment of some carbons in our MBOA molecule showed that it was necessary to take account of the long-range, carbon-proton couplings. Focusing at first on the methine carbons, C(4), C(6), and C(7) (5-MBOA),C(4), C(5), and C(7) (6-MBOA),and C(4), C(5), and C(6) (7-MBOA) (Figure 11,we found that all are or doublets of doublets (~JcH and 3JCCCH). doublets (lJC~) In each of the three cases, only one methine carbon exhibited no further splitting (no 'JCCCH), which fits respectively C(7) (5-MBOA), C(4) (6-MBOA), and C(5) (7MBOA), each having no vicinal protons. The remaining methine signals show a long-range coupling constant (3JCmH) (Table 11). The 'JCCH for those methine carbons is too small (1-2 Hz), in contrast to the vicinal coupling , J c C c H (7-12 Hz) with the exception of heterocyclic aromatic compounds, where the geminal coupling can also become large according to Weigert and Roberta (1968). In our molecule, both rules can apply since the methine carbons are not in the heterocyclic aromatic ring. The quaternary carbons C(8) and C(9) for 5-, 6-, and 7-MBOA each have different patterns depending on whether they have one or two vicinal protons. The spectrum obtained for 5-MBOA clearly reveals a symmetric seven-line pattern for the resonance at 137.52 ppm from which coupling constants can be extracted that are com~ 8.2 Hz, 3Jccc= ~ 7.7 Hz, patible only with C(8): 3 J c c c = and 2 J c c H = 3.6 Hz. C(9) of 5-MBOA has only one vicinal proton, and in this case, the pattern can only be a doublet of a doublet (see Table I1 for the value). For 6-MBOA,
5-MBOA C,H,
165.3
lJCCH
C,H,
4.5
3JCCCH
C,H,
5,614.7
C6H6
163.25 C,H,
C 3 7 7.9
C,H,
166.90 C,H,
C6H4
4.9 7.7
C9H4 C.H,
6-MBOA C,H, C,H, C,H,
165.90 C,H, 162.20 C,H, 165.25
4.5 4.1
7-MBOA C,H, C,H, C,H6
167.80 C,H, 163.00 162.70
3.2
C,H, C,H, C,H, C,H,
PZO) 6.9 5.014.7 8.7 9.75
C,H, 8.018.2 C,H, 8.017.7 C,H, 8.0 C,H, 7.9 C,H, 9.2 Table 111. Biological Evaluations of 5-, 6-, and 7-MBOA compounds bioassay type
5-MBOA 6-MBOA 7-MBOA
radish seedling 10 25 10 tomato seedline 10 25 10 P. maltophilia Ob t t c Southwestern corn borer, 0.20d 0.40 0.40 ED,,, % fall armyworm, ED,,, % >0.50 >0.50 0.40 European corn borer, 0.05 0.12 0.175 ED,,, 7% a Growth in 100 ppm of test compound; percent of control. Not active. ' 100 f i g applied to a bacteriological test disk paper produced in 10-mm clear zone o n a nutrient agar inoculated with the test bacteria. ED,, = percent in diet required to reduce larval growth by 50%.
the situation is reversed for C(8) and C(9), even if C(9) does not present the complete pattern of seven lines. Carbons 8 and 9 of 7-MBOA have the same shift when the spectrum is recorded in MezSO solution. The pattern, therefore, after gated decoupling is quite complicated and appears as d X d with another doublet under the d X d lines, but it is still possible to determine the coupling constant (see Table 11). The spectrum of 7-MBOA was also obtained in CDC1, to resolve C(8) and C(9). The quaternary carbons bearing the methoxyl group C(5), C(6), and C(7) appear differently. C(5) (5-MBOA) gives a doublet of a doublet as well as C(6) (6-MBOA)even if this last one [C(6)] is hidden by C(2). Carbon 7 (7MBOA) appears as an unresolved multiplet and is more shielded than the same carbons bearing a methoxyl group for 5- and 6-MBOA. This is due to the mesomeric effect of the oxygen of the five-membered ring. This effect is also observed for C(8) in 7-MBOA and for (8) in 5-MBOA. The three compounds inhibited the growth of radish (Raphanus satiuus) and tomato (Lycopersicon esculentum) seedlings, but only 6-MBOA was effective in reducing the growth to 15% of the control at 50 ppm (see Table 111). Both 7-MBOA and BMBOA were inactive at 50 ppm. The inhibition of growth has been attributed by Venis and Watson (1978) to the inhibition of auxins. Only 6-MBOA inhibited the growth of the bacterium P. maltophilia in an agar plate bioassay. When incoroporated in laboratory diets, 5- and 6-MBOA inhibited the larval growth of three insects, the Southwestern corn borer, the fall armyworm, and the European corn borer.
J. Agrlc.
Food Chem. 1982, 30, 1135-1140
ACKNOWLEDGMENT We acknowledge the excellent technical support of Thomas Oswalt and Susan Waage. LITERATURE CITED Chen, C. M.; Chen, T. Phytochemistry 1976,15, 1997. Davis, F. M.; Hedin, P. A. Crop Science Research Laboratory and Chemistry Research Unit, USDA-ARS, Mississippi Stab, MS 39762,unpublished data, 1982. Gahagan, H. E.; Mumma, R. 0. Phytochemistry 1967,6,1441. Klun, J.; Robinson, J. F. J. Econ. Entomol. 1969,62, 214. Matsumoto, I.; Okazawa, M. Japanese Patent 74 31 663,1974. Nicollier, G.; Thompson, A. C. J . Agric. Food Chem. 1982,30, 760.
1135
Richey, D. J.; Caskey, A. L.; Bemiller, J. N. Agric. Biol. Chem. 1976,40,2413. Venis, M. A.; Watson, P. J. Planta 1978,142, 103. Virtanen, A. I.; Hictala, P. K. Acta Chem. Scand. 1955,9,1543. Weigert, F.J.; Roberts, J. D. J . Am. Chem. SOC.1968,90,3543. Received for review March 18,1982. Accepted July 19, 1982. Mention of a trademark,proprietary product, or vendor does not constitute a guarantee or warranty of the product by the U S . Department of Agriculture or Mississippi State University and does not imply its approval to the exclusion of other products or vendors that may also be suitable.
Multiple Discriminant Analysis in the Analytical Differentiation of Venetian Wines. 3. A Reelaboration with Addition of Data from Samples of 1979 Vintage Prosecco Wine Giuseppe Scarponi,* Ivo Moret, Gabriele Capodaglio, and Paolo Cescon
Nineteen samples of Prosecco white wine, vintage of 1979, were analyzed for sodium, potassium, calcium, magnesium, chloride, pH, total acidity, phosphorus, ash content, and alkalinity of the ashes. Some parameters showed relevant differences with respect to the 1977 data. The new data were added to that already in our file, and the multiple discriminant analysis for the differentiation of the white wines Soave Classico, Prosecco di Conegliano-Valdobbiadene, and Verduzzo del Piave was applied. The apparent error rate obtained with all the parameters in the analysis was about 8.2%. Classification was improved (apparent error rate 6.6%) when the parameters calcium, alkalinity of the ashes, and ash content were excluded from the analysis. In this case the expected actual error rate, obtained by the jackknife procedure, was 9.8%. The number of parameters to be used can be reduced to five, without an error increment, or to four, with an apparent error rate of about 10%.
In previous papers, it was shown that the problem of differentiation of wines by means of analytical parameters can be successfully dealt with by statistical treatment of the experimental data according to the method of discriminant analysis (Moret et al., 1980; Scarponi et al., 1981). In such a method, the parameters are combined into one or more linear functions (the discriminant functions), the coefficients of which are computed to maximize the separation between the groups of different wine samples and to minimize the spread within the groups. Once the discriminant functions are obtained, it is possible to calculate values of them for each sample examined (discriminant scores) and, from these, to carry out the differentiation and the classification of the wines. The optimal combination of several parameters allows one to obtain a better exploitation of discriminant information contained in the data and to gain real means of differentiating the wines according to their origin. Previous treatments of experimental data concerning the Venetian white wines Soave Classico, Prosecco di Conegliano-Valdobbiadene, and Verduzzo del Piave, vintage of 1977, showed that the three wines could be differentiated both in two by two comparisons, by binary discriminant analyses (Moret et al., 1980), and in a ternary comparison, by multiple discriminant analysis (Scarponi et al., 1981). The following analytical determinations were utilized as discriminating parameters: sodium, potassium, calcium, Cattedra di Chimica Analitica, Facolth di Chimica Industriale, Universitii di Venezia, 1-30123 Venezia, Italy. 0021-8561 18211430-1135$01.25/0
magnesium, chloride, pH, total acidity (TA), phosphorus, ash content (AC), and alkalinity of the ashes (AA). The values of the selected parameters can, however, change significantly from year to year, especially as a consequence of climatic and environmental factors. Then, to obtain the discriminant functions that can be valid through the years, it is necessary to introduce into the analysis the data obtained from several years, so as to weigh the parameters according to their variability through the years. In this paper we report the results of the analytical measurements mentioned above and performed on samples of Prosecco wine, vintage of 1979. Some significant differences from the previously obtained results (vintage of 1977) are evidenced. The new data were added to that already in our file, and the multiple discriminant analysis was repeated. Discriminant and classification function coefficients are reported, and the differentiation of the wines is evaluated. EXPERIMENTAL SECTION Collection and Analysis of Samples. Nineteen samples of Prosecco wine, vintage of 1979, were collected in the production zone from lots for which genuineness and typicality were guaranteed and certified by the Italian D.O.C. (Denominazionedi Origine Controllata-Certified Brand of Origin) brand. This brand is obtained only by producers that follow the production disciplinary rules (“Gazzetta Ufficiale della Repubblica Italiana”, 1969). These rules prescribe that, in order to be classified as D.O.C., the wine, in addition to being judged by its orga0 1982 American Chemical Society
